| 证券代码：300454 | 证券简称：深信服 | 公告编号：2025-079 | | --- | --- | --- | | 债券代码：123210 | 债券简称：信服转债 | | 深信服科技股份有限公司 关于变更注册资本、公司类型并修订 《公司章程》及其附件的公告 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 深信服科技股份有限公司（以下简称"公司"）于 2025 年 12 月 9 日召开公司 第三届董事会第三十四次会议，审议通过了《关于变更注册资本、公司类型并修 订〈公司章程〉及其附件的议案》。本议案尚需提交公司股东会审议。现将具体 情况公告如下： 一、本次注册资本及股份总数拟变更情况 除上述内容外，《公司章程》的其他条款不变。本次《公司章程》的修订尚 需提交公司股东会审议，公司董事会授权公司相关人员在股东会审议通过后办理 工商变更登记等相关事宜，本次《公司章程》条款的最终修改情况以市场监督管 2 / 3 理部门的审批结果为准。 公司因注销已回购的股份 1,255,715 股，致公司总股本由 421,883,707 股变更 为 420,627,992 ...

| 证券代码：300454 | 证券简称：深信服 | 公告编号：2025-080 | | --- | --- | --- | | | 债券简称：信服转债 | | | 债券代码：123210 | | | 深信服科技股份有限公司 关于补选第三届董事会非独立董事的公告 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 深信服科技股份有限公司 董事会 二〇二五年十二月九日 1 / 2 附件：第三届董事会非独立董事候选人简历 肖立业，男，1987 年出生，中国国籍，无永久境外居留权，本科毕业于广 东外语外贸大学，硕士毕业于香港中文大学（深圳）。2011 年至 2014 年担任广 东宜通世纪科技股份有限公司审计主管，2014 年至今任本公司内审部主管，于 2016 年至 2025 年兼任公司职工代表监事。 深信服科技股份有限公司（以下简称"公司"）于 2025 年 12 月 9 日召开第三 届董事会第三十四次会议，审议通过了《关于补选第三届董事会非独立董事的议 案》，现将具体情况公告如下： 为进一步优化公司治理结构，提高董事会决策的科学性、有效性，根据《中 华人民共和国公司 ...

| 证券代码：300454 | 证券简称：深信服 | 公告编号：2025-081 | | --- | --- | --- | | 债券代码：123210 | 债券简称：信服转债 | | 深信服科技股份有限公司 关于召开 2025 年第二次临时股东会的通知 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 一、召开会议的基本情况 1、股东会届次：2025 年第二次临时股东会 2、股东会的召集人：董事会 3、会议召开的合法、合规性： （1）本次股东会召开符合《公司法》《证券法》《上市公司股东会规则》 等相关法律、法规和《公司章程》的有关规定。 （2）公司第三届董事会第三十四次会议审议通过了《关于召开 2025 年第二 次临时股东会的议案》。 4、会议时间： 7、出席对象： （1）截至 2025 年 12 月 17 日下午收市时在中国证券登记结算有限责任公司 深圳分公司登记在册的公司全体股东。前述公司全体股东均有权出席本次股东会， 不能亲自出席现场会议的股东可以以书面形式委托代理人出席会议和参加表决 1 / 7 （该股东代理人不必是公司股东）； 8、会议地点：广东省深 ...

| 证券代码：300454 | 证券简称：深信服 | 公告编号：2025-078 | | --- | --- | --- | | | 债券简称：信服转债 | | | 债券代码：123210 | | | 深信服科技股份有限公司 第三届董事会第三十四次会议决议公告 本公司及董事会全体成员保证信息披露的内容真实、准确、完整，没有虚假 记载、误导性陈述或重大遗漏。 一、董事会会议召开情况 深信服科技股份有限公司（以下简称"公司"）于 2025 年 12 月 5 日以电子邮 件等方式发出会议通知，于 2025 年 12 月 9 日在公司会议室以现场及通讯方式召 开第三届董事会第三十四次会议。本次会议应出席董事 6 名，实际出席董事 6 名。本次董事会会议由董事长何朝曦先生主持。本次会议的召集和召开符合《公 司法》等法律法规、规范性文件以及《公司章程》的规定。 二、董事会会议审议情况 本议案已经董事会提名委员会审议通过。 具体内容详见公司在巨潮资讯网上披露的公司《关于补选第三届董事会非独 立董事的公告》（公告编号：2025-080）。 经与会董事审议，以投票表决方式通过了以下决议： (一)审议通过《关于变更注册资本、公司 ...

Core Viewpoint - The article highlights the urgent need for advanced security mechanisms in the face of rapidly evolving AI-driven attacks, emphasizing that traditional passive defenses are inadequate against modern threats [3][5][6]. Group 1: Security Challenges - A significant reduction in the average time required for successful attacks has been observed, decreasing from 9 days in 2021 to just 25 minutes in 2023 [3]. - The emergence of AI-driven attacks allows hackers to simulate legitimate behavior, bypassing traditional security measures and dynamically adapting to defenses [6]. - The current state of security in enterprises is characterized by fragmented defenses, where companies invest heavily in hardware but neglect ongoing operational security, leading to greater vulnerabilities [6]. Group 2: Need for Proactive Security - There is a critical need for security to be integrated into the design of technology from the outset, rather than relying on post-incident patches [5][6]. - The concept of "embodied intelligence" in robotics necessitates a focus on security as a prerequisite for widespread adoption in production and daily life [5]. - The industry is urged to adopt a collaborative mindset and conduct tailored drills to enhance security preparedness [5]. Group 3: AI in Security Defense - AI technologies are being leveraged to significantly enhance security operations, with systems capable of processing vast amounts of data and reducing false alarms by 90% [8]. - New security architectures are emerging that utilize AI for autonomous decision-making and response, aiming to create a more adaptive defense against threats [8]. - The challenge remains that attackers are also employing AI to refine their strategies, necessitating a shift in how defenses are conceptualized [8]. Group 4: Security Insurance and Accountability - The article advocates for the promotion of cybersecurity insurance, which would assess the effectiveness of security measures and provide compensation based on actual defense outcomes [10][11]. - Recent initiatives by regulatory bodies aim to integrate cybersecurity insurance into the broader risk management framework for enterprises, with pilot programs already showing significant uptake [10][11]. - The future of security may hinge on the ability to create verifiable and sustainable operational mechanisms, moving away from compliance-based approaches to those focused on demonstrable capability [11].

Group 1 - The core issue highlighted is the increasing vulnerability of security systems in the face of AI-driven attacks, with the average time to successfully execute an attack decreasing from 9 days in 2021 to just 25 minutes in 2023 [1] - The GEEKCON competition showcased a significant security flaw in a humanoid robot, allowing attackers to remotely control it through a voice command, which raises concerns about systemic risks in future robotic clusters [2] - There is a pressing need for security mechanisms to be integrated from the design phase, rather than relying on post-incident patches, as many companies currently focus on compliance rather than effective security measures [3] Group 2 - The current approach to security, characterized by fragmented defenses and reactive measures, is ineffective against AI-driven threats, as attackers can now simulate legitimate behavior to bypass security systems [4] - The introduction of AI in security operations has the potential to drastically improve efficiency, with AI systems capable of processing significantly more data compared to manual methods, thus enhancing risk monitoring [6] - New security architectures are emerging, such as those proposed by companies like Palo Alto Networks and Fortinet, which aim to create adaptive and self-evolving security systems [6] Group 3 - The concept of pricing security based on effectiveness rather than compliance is gaining traction, with calls for the promotion of cybersecurity insurance to alleviate user anxiety and assess the true capabilities of security vendors [7] - Recent initiatives by the Chinese government to promote cybersecurity insurance indicate a shift towards integrating financial services with cybersecurity, aiming to enhance corporate risk management capabilities [7][8] - The future of cybersecurity may depend on the establishment of verifiable and sustainable operational mechanisms, as insurance models could incentivize companies to improve their defensive capabilities [8]

Group 1 - The core viewpoint of the news highlights a significant surge in the optical module CPO industry chain, with the AI index on the ChiNext board rising by 4.94% and key stocks like Tianfu Communication and Guangku Technology experiencing substantial gains [1][2] - The ChiNext AI ETF, Huaxia (159381), has seen a rapid increase in investment, attracting over 100 million yuan in the last five days, reaching a new high of over 1.1 billion yuan in total assets [1] - The AWS re:Invent cloud computing conference showcased advancements in AI training chips, including the introduction of Trainium 4, which supports NVLink Fusion technology for enhanced performance with NVIDIA GPUs [1] Group 2 - Guosen Securities indicates an accelerated global AI arms race in cloud service providers (CSP), benefiting the computing power infrastructure, and recommends focusing on optical devices, optical modules, communication equipment, liquid cooling, and edge computing [2] - As of November 28, 2025, the top ten weighted stocks in the ChiNext AI index account for 64.98% of the index, with notable companies including Zhongji Xuchuang and Xinyi Sheng [2] - The ChiNext AI ETF, Huaxia, has associated products for investors, including the Huaxia ChiNext AI ETF Initiated Link A and C [2]

Investment Rating - The report maintains a "Recommend" rating for the industry, indicating a positive outlook for the sector's fundamentals and expected performance relative to the benchmark index [3]. Core Insights - The report emphasizes the importance of AI applications, highlighting the competitive landscape for AI operating systems as companies like Alibaba and ByteDance launch new products [6][21]. - It suggests a strategic focus on the fintech sector, particularly in light of recent regulatory changes that may encourage insurance companies to increase market participation [27][28]. Summary by Sections Industry Overview - The industry comprises 286 stocks with a total market capitalization of 4220.9 billion and a circulating market value of 3742.6 billion [3]. - Recent performance metrics show a 1-month absolute performance of -4.3%, a 6-month performance of 17.2%, and a 12-month performance of 16.4% [5]. Recent Developments - ByteDance launched the "Doubao" mobile assistant, enhancing user interaction through AI capabilities [9][11]. - DeepSeek introduced two new models, DeepSeek-V3.2 and DeepSeek-V3.2-Speciale, which exhibit leading inference capabilities in the global market [16][20]. - Amazon unveiled its new AI training chip, Trainium 3, aimed at competing with Nvidia and Google in the AI chip market [21][22]. Market Performance Review - The computer sector experienced a decline of 1.73% in the first week of December 2025, with notable gainers including Hangtian Zhizhuang and Dineike [29][30].

从估值层面来看，创业板成长ETF跟踪的创成长最新市盈率（PE-TTM）仅38.18倍，处于近10年 39.00%的分位，即估值低于近10年61.00%以上的时间，估值适中。 建议逢低关注。 截至2025年12月5日 10点44分 ，A股三大指数涨跌互现，上证指数跌0.15%，深证成指涨0.08%，创 业板指涨0.14%。创业板成长ETF（159967）上涨0.51%，最新报价0.596元，盘中成交额1.1亿元，换手 率3.41%。成分股方面涨跌互现，太辰光领涨8.12%，光库科技上涨6.38%，天孚通信上涨5.16%，惠城 环保上涨3.31%，锐科激光上涨2.41%；鼎龙股份领跌1.45%，深信服下跌1.27%，全志科技下跌1.23%， 胜宏科技下跌1.07%，汇川技术下跌0.78%。 从近期表现来看，创业板成长ETF走势明显强于创业板指数近8个交易日上涨8.81%，同期创业板指 数上涨4.73%。 | 涨跌幅 | | | 振唱 | (VI /7 / | | --- | --- | --- | --- | --- | | 159967(创业板成长) . | 0.0 | 0.048 | | % 1298.20% ...

Core Viewpoint - The article discusses the performance of the ChiNext Artificial Intelligence ETF managed by Huaxia Fund Management, highlighting its recent market activity and returns since inception [1]. Group 1: ETF Performance - The ChiNext Artificial Intelligence ETF (159381) opened with a gain of 0.53%, priced at 1.695 yuan [1]. - Since its inception on March 14, 2025, the fund has achieved a return of 68.77%, with a monthly return of 3.20% [1]. Group 2: Top Holdings - Key stocks in the ETF include: - Zhongji Xuchuang, which opened up by 0.75% - Xinyi Sheng, which increased by 1.98% - Tianfu Communication, which rose by 3.14% - Xiechuang Data, which gained 0.55% - Ruantong Power, which went up by 0.87% - Allwinner Technology remained unchanged at 0.00% - Shenxinfu increased by 0.10% - Beijing Junzheng rose by 0.61% - Zhongke Chuangda increased by 0.31% - Jingjiawei decreased by 0.39% [1].